About the Why? and the How? of psychologically plausible agents

Andreas Ernst

December 5, 2012


To deepen the analysis and understanding of human behaviour by using artificial societies, a credible modelling of agents and their decision processes is needed. Such a deep analysis can be informed by the psychological knowledge about processes and triggers of specific behaviours, given the interaction between an agent’s environment and some individual preference set. This requires a framework capable of handling a high number of deep, heterogeneous agents, together with their individual networks.

In this presentation, a software framework is presented that aims at providing psychological plausibility to the modelling of so-called citizen agents. It fills the gap between (flat) agent frameworks without any built-in psychological foundations on the one hand and full-fledged (deep) cognitive architectures on the other hand, the latter being computationally too intensive to be of any use in larger scale problems. The framework provides prefabricated components of an agent’s decision process like perception, memory, different modes of decision making, basic learning algorithms, or social influence. Each of these components is based on appropriate psychological empirical results or theories.

Using this framework, together with specifically gathered data, an agent’s interaction with other agents, technology, or the natural environment can be investigated. Examples include energy use, innovation adoption, or helper networks. Besides the emerging macro phenomena, their micro foundation lying in the individual preferences and processes are important data produced by the framework. Individual, local perception of the technical or physical environment and the information of the individual social network are crucial in determining an agent’s behaviour.

To model larger amounts of social data, with a higher number of agents together with their networks, the sociological concept of lifestyles can be used as a classifying element, and as a bridge to upscale the individual data to the agent set.


Prof. Dr. Andreas Ernst achieved his psychology Diplom degree with a focus on psychology of knowledge/cognitive science in 1988. He finished his PhD—entitled "Social knowledge as basis of decisions in conflict situations"—in 1993. Until 1999 he was assistant professor at the Institute of Psychology, University of Freiburg i. Br. and in the same year he finished his postdoctoral lecture qualification under the title: "Information dilemmas and the use of natural resources". After another three years at the Institute of Psychology, University of Freiburg i.Br. he became full professor for Environmental Systems Analysis at the Center for Environmental Systems Research of the University of Kassel. He also heads the Graduate Program "Man-Environment-Systems" (ProMUS) and is the current president of the European Social Simulation Association (ESSA). Prof. Dr. Andreas Ernst is co-editor of the book series "Social Science Simulations" of the Metropolis-Verlag.

Personal Website

Big Data and the Attention Economy

Bernardo Huberman

December 6, 2012


The advent of the web has led to an ever expanding ocean of data whose outlines are hard to discern but its effects are easy to feel. It reflects a societal change that is both global and daunting at the same time. Global because it encompasses content created by people seeking attention and information from all over the world without much awareness of privacy issues. And daunting since it is hard to discern what to pay attention to when confronted with such flood of content.

This talk will describe the effects that the attention economy has on the production and consumption of big data and the opportunities that big data offers about learning the behavior of large groups of people and the design of novel socially attentive systems. It will also address the questions that big data poses about privacy and how markets for private data can restore a sense of control to most users of the web.


Bernardo A. Huberman is a Senior HP Fellow and director of the Social Computing Research Group at HP Labs, which focuses on methods for harvesting the collective intelligence of groups of people in order to realize greater value from the interaction between users and information. Huberman’s main research focus is on the relationship between local actions and the global behavior of large, distributed systems. Areas of exploration include distributed knowledge, social organizations and the economics of attention. Much of Huberman's research has concentrated on the World Wide Web, with an emphasis on the dynamics of its growth and use. This work helped uncover the nature of electronic markets, the detailed structure of the web and the laws governing the way people surf for information. One of the originators of the field of ecology of computation, Huberman recently published the book, "The Laws of the Web: Patterns in the Ecology of Information," with MIT Press.

Personal Website

From Computational Social Science to Socio-Inspired Technology to Artificial Societies

Dirk Helbing

December 7, 2012


What can we learn from the way society is organized? What are the underlying principles? How can we use them to design new technological systems?

Along the lines of these guiding questions, I will start with models of pedestrians and crowds, and the applications in logistics and traffic light control they have inspired.

It will be shown that self-organization is a wide-spread principle underlying the emergence of social coordination, cooperation, and social norms. These phenomena can now be dynamically modeled based on evolutionary principles, which are transferable to ICT systems as well. I will also discuss, how and why the wrong kinds of system designs can lead to breakdowns of traffic flows, cooperation, or financial markets. It will be argued that the theory of complex systems can help to provide an explanatory understanding of desirable and undesirable cascading effects and guidelines for the design of socially interactive systems (such as the 'ecosystem' of financial trading algorithms).


Dirk Helbing has worked as Managing Director of the Institute for Transport & Economics at TU Dresden and is now Professor of Sociology, in particular of Modeling and Simulation at ETH Zurich. Having studied physics and mathematics, he investigates complex social, economic, and transport systems with methods from statistical physics, individual-based models, and behavioral experiments. Helbing is well-known for the social force model, in particular its application to self-organization phenomena in pedestrian crowds. Besides the slower-is-faster effect, he introduced the freezing-by-heating effect and the phase diagram of congested traffic states. Recent work applies principles of collective intelligence and dynamics to the optimization of freeway and urban traffic flows. In game theory, Helbing proposed a microscopic foundation of evolutionary game theory and studied self-organized behavioral conventions early on. His current work develops socio-inspired technologies and investigates the role of success-driven motion for the establishment of cooperation among selfish individuals.

Personal Website


top

Full Papers

Misita Anwar and Graeme Johanson Mobile Phones, Family and Personal Relationships: the case of Indonesian Micro-entrepreneurs
Maartje Groenewegen, Dimo Stoyanov, Dirk Deichmann and Aart van Halteren Connecting with Active People Matters: The Influence of an Online Community on Physical Activity Behavior Best Paper Award
Zhu Wang, Daqing Zhang, Dingqi Yang, Zhiyong Yu and Xingshe Zhou Detecting Overlapping Communities in Location-Based Social Networks
Xiaoyue Ma and Jean-Pierre Cahier Collaboratively Constructing a VDL-based Icon System for Knowledge Tagging
Wynn Sterling, Christophe Giraud-Carrier and Teppo Felin A Framework for the Design and Synthesis of Coordinated Social Systems
Sasan Yazdani, Ivan Ivanov, Morteza Analoui, Reza Berangi and Touradj Ebrahimi Spam Fighting in Social Tagging Systems
Bogdan Gliwa, Jaroslaw Kozlak, Anna Zygmunt and Krzysztof Cetnarowicz Models of social groups in blogosphere with information about comment addressees and sentiments Best Paper Award
Piotr Bródka, Przemyslaw Kazienko and Bartosz Kołoszczyk Predicting Group Evolution in the Social Network
Peter Laflin, Alexander Mantzaris, Peter Grindord, Fiona Ainley, Amanda Otley and Desmond Higham Dynamic Targeting in an Online Social Medium
Marco Conti, Andrea Passarella and Fabio Pezzoni A Model to Represent Human Social Relationships in Social Network Graphs
Pnina Fichman How Many Answers Are Enough? Optimal Number of Answers for Q&A Sites
Patrick Minder and Abraham Bernstein CrowdLang: A Programming Language for the Systematic Exploration of Human Computation Systems
Lu Chen, Wenbo Wang and Amit Sheth Are Twitter Users Equal in Predicting Elections? A Study of User Groups in Predicting 2012 U.S. Republican Presidential Primaries
Paul de Laat Navigating Between Chaos and Bureaucracy: How Open-Content Communities are Backgrounding Trust
Zhenzhen Zhao, Xiaodi Huang and Noel Crespi A System for Web Widget Discovery Using Semantic Distance between User Intent and Social Tags
Hatem Ghorbel Experiments in Cross-Lingual Sentiment Analysis in Discussion Forums
Jaroslaw Jankowski, Radosław Michalski and Przemyslaw Kazienko The Multidimensional Study of Viral Campaigns as Branching Processes
Julie Birkholz, Rena Bakhshi, Ravindra Harige, Peter Groenewegen and Maarten Van Steen Scalable Analysis of Socially Informed Network Models
Milen Chechev and Petko Georgiev A multi-view content-based user recommendation scheme for following users in Twitter
Rob Duell and Jan Treur A Computational Analysis of Joint Decision Making Processes
Su Mon Kywe, Ee-Peng Lim and Feida Zhu A Survey of Recommender Systems in Twitter

Short Papers

Ansuya Ahluwalia, Allen Huang, Roja Bandari and Vwani Roychowdhury An Automated Multiscale Map of Conversations: Mothers and Matters
Ward Van Breda, Jan Treur and Arlette Van Wissen Analysis and Support of Lifestyle via Emotions Using Social Media
Thomas Paul, Martin Stopczynski, Daniel Puscher, Melanie Volkamer and Thorsten Strufe C4PS - Helping Facebookers Manage their Privacy Settings
Su Mon Kywe, Tuan-Anh Hoang, Ee-Peng Lim and Feida Zhu On Recommending Hashtags in Twitter Networks
Zeinab Abbassi, Christina Aperjis and Bernardo Huberman Swayed by Friends or by the Crowd?
Francois Bavaud and Guillaume Guex Interpolating between Random Walks and Shortest Paths: a Path Functional Approach
John Pfaltz Entropy in Social Networks
Jaroslaw Jankowski, Sylwia Ciuberek, Anita Zbieg and Radosław Michalski Studying Paths of Participation in Viral Diffusion Process within Virtual Chat Environment
Somya Joshi, Timo Wandhoefer, Vasilis Koulolias, Catherine Van Eeckhaute, Beccy Allen and Steve Taylor Paradox of Proximity – Trust & Provenance within the context of Social Networks & Policy
Norhidayah Azman, David Millard and Mark Weal Dark Retweets: Investigating Non-Conventional Retweeting Patterns
Christopher Fröch and Martin Schumann Quality assessment of user comments on mobile platforms considering channel of activation and platform design
Andrzej M.J. Skulimowski A Foresight Support System to Manage Knowledge on Information Society Evolution
Barbara Carminati, Elena Ferrari and Marco Viviani A Multi-dimensional and Event-based Model for Trust Computation in the Social Web
Guillermo De Ita, Luis Altamirano, Aurelio Lopez and Yolanda Moyao A Method Based on Congestion Game Theory for Determining Electoral Tendencies
Imrul Kayes How Influential are You: Detecting Influential Bloggers in a Blogging Community
Somprakash Bandyopadhyay, Apratim Mukherjee and Shrabastee Banerjee A Simulation Model using Transaction Cost Economics to Analyze the Impact of Social Media on Online Shopping
Jean-Pierre Cahier, Nour El Mawas and Aurélien Bénel Dynamic "Participative Rules" in Serious Games, New Ways for Evaluation?
Markus Schaal, John O'Donovan and Barry Smyth An Analysis of Topical Proximity in the Twitter Social Graph
Soheila Abrishami, Mahmoud Naghibzadeh and Mehrdad Jalali Web Page Recommendation Based on Semantic Web Usage Mining

Demo and Posters

Folke Mitzlaff and Gerd Stumme Namelings - Discover Given Name Relatedness Based on Data from the Social Web
Maurizio Tesconi, Davide Gazzè and Angelica Lo Duca SocialTrends: a Web Application for Monitoring and Visualizing Users in Social Media
Peter Laflin, Desmond Higham, Fiona Ainley, Amanda Otley and Alex Mantzaris Demonstration of Dynamic Targeting in an Online Social Medium

top
08:00 Registration opens
08:45 Conference opening
09:00
Keynote Talk 1: About the Why? and the How? of psychologically plausible agents
Speaker: Andreas Ernst, University of Kassel
Session Chair: Karl Aberer, EPFL
10:00 Break
10:30
Research Session 1: Social Graph, Social Influence and Viral Marketing
Session Chair: Ee-Peng Lim
Full talks
(25 min)
Connecting with Active People Matters: The Influence of an Online Community on Physical Activity Behavior
Maartje Groenewegen, Dimo Stoyanov, Dirk Deichmann and Aart van Halteren
The Multidimensional Study of Viral Campaigns as Branching Processes
Jaroslaw Jankowski, Radosław Michalski and Przemyslaw Kazienko
A Model to Represent Human Social Relationships in Social Network Graphs
Marco Conti, Andrea Passarella and Fabio Pezzoni
Short talks
(15 min)
Interpolating between Random Walks and Shortest Paths: a Path Functional Approach
Francois Bavaud and Guillaume Guex
Studying Paths of Participation in Viral Diffusion Process within Virtual Chat Environment
Jaroslaw Jankowski, Sylwia Ciuberek, Anita Zbieg and Radosław Michalski
How Influential are You: Detecting Influential Bloggers in a Blogging Community
Imrul Kayes
Dark Retweets: Investigating Non-Conventional Retweeting Patterns
Norhidayah Azman, David Millard and Mark Weal
12:30 Lunch break
14:00
Research Session 2: Recommendation and Crowd Computing
Session Chair: Pnina Fichman
Full talks
(25 min)
A Framework for the Design and Synthesis of Coordinated Social Systems
Wynn Sterling, Christophe Giraud-Carrier and Teppo Felin
CrowdLang: A Programming Language for the Systematic Exploration of Human Computation Systems
Patrick Minder and Abraham Bernstein
A multi-view content-based user recommendation scheme for following users in Twitter
Milen Chechev and Petko Georgiev
A Survey of Recommender Systems in Twitter
Su Mon Kywe, Ee-Peng Lim and Feida Zhu
Short talks
(15 min)
Swayed by Friends or by the Crowd?
Zeinab Abbassi, Christina Aperjis and Bernardo Huberman
Quality assessment of user comments on mobile platforms considering channel of activation and platform design
Christopher Fröch and Martin Schumann
16:00 Break
16:30
Tutorial Session 1

Online Social Experiments with nodeGame
Stefano Balietti, ETH Zürich
18:00 End of scientific program
09:00
Keynote Talk 2: Big Data and the Attention Economy
Speaker: Bernardo Huberman, HP Research
Session Chair: Sabine Süsstrunk
10:00 Break
10:30
Research Session 3: Sentiment Analysis and Trust
Session Chair: Amit Sheth
Full talks
(25 min)
Experiments in Cross-Lingual Sentiment Analysis in Discussion Forums
Hatem Ghorbel
Models of social groups in blogosphere with information about comment addressees and sentiments
Bogdan Gliwa, Jaroslaw Kozlak, Anna Zygmunt and Krzysztof Cetnarowicz
Navigating Between Chaos and Bureaucracy: How Open-Content Communities are Backgrounding Trust
Paul de Laat
Short talks
(15 min)
Analysis and Support of Lifestyle via Emotions Using Social Media
Ward Van Breda, Jan Treur and Arlette Van Wissen
An Automated Multiscale Map of Conversations: Mothers and Matters
Ansuya Ahluwalia, Allen Huang, Roja Bandari and Vwani Roychowdhury
Paradox of Proximity - Trust & Provenance within the context of Social Networks & Policy
Somya Joshi, Timo Wandhoefer, Vasilis Koulolias, Catherine Van Eeckhaute, Beccy Allen and Steve Taylor
A Multi-dimensional and Event-based Model for Trust Computation in the Social Web
Barbara Carminati, Elena Ferrari and Marco Viviani
C4PS - Helping Facebookers Manage their Privacy Settings
Thomas Paul, Martin Stopczynski, Daniel Puscher, Melanie Volkamer and Thorsten Strufe
13:00
Buffet lunch with poster/demo session
14:00
Research Session 4: Social Tagging and Discovery
Session Chair: Andreas Ernst
Full talks
(25 min)
A System for Web Widget Discovery Using Semantic Distance between User Intent and Social Tags
Zhenzhen Zhao, Xiaodi Huang and Noel Crespi
Dynamic Targeting in an Online Social Medium
Peter Laflin, Alexander Mantzaris, Peter Grindord, Fiona Ainley, Amanda Otley and Desmond Higham
Spam Fighting in Social Tagging Systems
Sasan Yazdani, Ivan Ivanov, Morteza Analoui, Reza Berangi and Touradj Ebrahimi
Collaboratively Constructing a VDL-based Icon System for Knowledge Tagging
Xiaoyue Ma and Jean-Pierre Cahier
Short talks
(15 min)
On Recommending Hashtags in Twitter Networks
Su Mon Kywe, Tuan-Anh Hoang, Ee-Peng Lim and Feida Zhu
Dynamic "Participative Rules" in Serious Games, New Ways for Evaluation?
Jean-Pierre Cahier, Nour El Mawas and Aurélien Bénel
16:00 Break
16:30
Tutorial Session 2

Supporting sociological theories with social media data mining
Luca Aiello, Yahoo! Research Barcelona, Spain
18:00 End of scientific program
09:00
Keynote Talk 3: From Computational Social Science to Socio-Inspired Technology to Artificial Societies
Speaker: Dirk Helbing, ETH Zürich
Session Chair: Andreas Flache
10:00 Break
10:30
Research Session 5: Community Detection and Evolution
Session Chair: Alois Ferscha
Full talks
(25 min)
Predicting Group Evolution in the Social Network
Piotr Bródka, Przemyslaw Kazienko and Bartosz Kołoszczyk
Scalable Analysis of Socially Informed Network Models
Julie Birkholz, Rena Bakhshi, Ravindra Harige, Peter Groenewegen and Maarten Van Steen
Detecting Overlapping Communities in Location-Based Social Networks
Zhu Wang, Daqing Zhang, Dingqi Yang, Zhiyong Yu and Xingshe Zhou
Short talks
(15 min)
An Analysis of Topical Proximity in the Twitter Social Graph
Markus Schaal, John O'Donovan and Barry Smyth
Entropy in Social Networks
John Pfaltz
Web Page Recommendation Based on Semantic Web Usage Mining
Soheila Abrishami, Mahmoud Naghibzadeh and Mehrdad Jalali
A Method Based on Congestion Game Theory for Determining Electoral Tendencies
Guillermo De Ita, Luis Altamirano, Aurelio Lopez and Yolanda Moyao
12:30 Lunch break
14:00
Research Session 6: Social Informatics and Applications
Session Chair: Michele Catasta
Full talks
(25 min)
Are Twitter Users Equal in Predicting Elections? A Study of User Groups in Predicting 2012 U.S. Republican Presidential Primaries
Lu Chen, Wenbo Wang and Amit Sheth
A Computational Analysis of Joint Decision Making Processes
Rob Duell and Jan Treur
How Many Answers Are Enough? Optimal Number of Answers for Q&A Sites
Pnina Fichman
Mobile Phones, Family and Personal Relationships: the case of Indonesian Micro-entrepreneurs
Misita Anwar and Graeme Johanson
Short talks
(15 min)
A Simulation Model using Transaction Cost Economics to Analyze the Impact of Social Media on Online Shopping
Somprakash Bandyopadhyay, Apratim Mukherjee and Shrabastee Banerjee
A Foresight Support System to Manage Knowledge on Information Society Evolution
Andrzej M.J. Skulimowski
16:00 Break
16:30
Plenary Panel: Making Big Data accessible for Research

Panel Moderator
Bernardo Huberman (HP Research)
Panelists
Dirk Helbing (ETH Zürich)
Frederic Kaplan (EPFL)
Thomas Hofmann (Google Zürich)

18:00 End of scientific program

top
4 December 2012
Room BC 420, EPFL

First international workshop on Just-in-time Sociology

JITSO 2012 will gather the most significant international researchers that try to understand social phenomena as they unfold, mining their digital traces. We would like to present studies focusing on large-scale social movements (“Arabic Spring Revolutions”, London riots, electoral campaigns, big controversies) as well as studies investigating new everyday social dynamics (analyses of photographic flux on Facebook, social dynamics in on-line games, etc.). The aim of the workshop is to draw a global map of the tools, methods and first results of this emerging field. One session will be dedicated to the presentation of data mining tools for analyzing streams of texts and images.

JITSO 2012 will be organized through an experimental open and public reviewing process, using WordPress as a central platform for publishing and reviewer articles and for designing the structure of the workshop itself.

Read more


Workshop on Internet Trust, Reputation, Identity and Privacy

Issues of individual trust in the use of personal data, the reputation of online services and users, and identity management tools are never far from the headlines. From the EU's revision of its data protection framework to the US National Strategy for Trusted Identities in Cyberspace, businesses and policymakers as well as academic researchers now have a firm interest in better understanding the underlying technical and social factors that underpin individual and organisational behaviour in this field.

EINS is therefore organising a one-day workshop to bring together university and industry researchers working on these topics. Co-located with SocInfo 2012 in Lausanne, it will have a particular focus on multidisciplinary perspectives that bring together researchers in the humanities and social sciences with mathematicians, engineering and physical scientists

Read more


Today, we know more about the universe than about our society. It's time to use the power of information to explore social and economic life on Earth and discover options for a sustainable future. Together, we can manage the challenges of the 21st century, combining the best of all knowledge.

The FuturICT Knowledge Accelerator is a previously unseen multidisciplinary international scientific endeavour with focus on techno-socio-economic-environmental systems.

The FuturICT meeting will have the following program:

14:00 The Growth of Information Networks
Matus Medo (University of Fribourg)
14:30 Collective Awareness - Models and Dynamics
Alois Ferscha (Institut für Pervasive Computing)
15:00 Checking In or Checked In: Comparing Large-Scale Location Disclosure Patterns
Daniel Gatica-Perez (EPF Lausanne)
15:30 The Structural Effects of Randomness on the Emergence of Social Norms (results from the laboratory)
Michael Mäs (ETH Zurich)


top